Recommended Articles
Legal & Policy

Texas Sues OpenAI and Seeks Court-Approved Safety Reviews for New AI Models

By Angeled · October 24, 2026 · Legal & Policy
Texas Sues OpenAI and Seeks Court-Approved Safety Reviews for New AI Models
Texas Sues OpenAI and Seeks Court-Approved Safety Reviews for New AI Models

On October 23, 2026, Texas Attorney General Ken Paxton sued OpenAI and its chief executive, Sam Altman, in state district court in Collin County, alleging that the company deceived Texans about the safety, accuracy, and reliability of ChatGPT. The case is the latest — and, in its requested remedies, one of the most far-reaching — in a fast-growing wave of state enforcement actions against generative AI products.

What Texas alleges

The 67-page petition brings claims under the Texas Deceptive Trade Practices Act and the SCOPE Act, a Texas law covering minors' online data and parental controls. Texas alleges that OpenAI marketed ChatGPT as safe and reliable while failing to protect users, including children, from foreseeable harms.

The state's account, as described in its announcement, lists alleged harms attributed to ChatGPT, including aiding mass shooters, encouraging suicidal behavior, causing public humiliation for professionals, eroding critical thinking, and addicting minors while collecting their data without parental oversight. These are allegations in a petition, not findings by a court, and OpenAI had not yet answered the suit at the time of reporting. They should be treated as claims to be tested in litigation, not as established facts about any individual incident.

Paxton framed the case as a duty question. "I will fight to protect Texans from the harms of AI and ensure that powerful technology companies are not above the law," he said, according to coverage of the filing.

What Texas is asking the court to do

The remedies Texas seeks go well beyond money, and that is why the case matters for the whole AI industry. Texas asks the court to bar OpenAI from offering ChatGPT to minors in the state, and to block any new OpenAI model for Texas users until a court-approved independent evaluator has reviewed its safety guardrails. Texas also seeks civil penalties of $10,000 per violation, disgorgement of assets, and a jury trial.

If granted, the model-review remedy would insert a court-supervised gate between an AI laboratory and its users in one of the largest states in the country. Release schedules, launches, and rollouts would have to be planned around an external review process, with timing and standards set outside the company.

Texas is the tenth state — and the pattern matters

Texas became the tenth state to sue OpenAI. Florida sued in June 2026, and eight states filed separate lawsuits on October 14, 2026, alleging that ChatGPT harms children and that the company misled parents about its safety. Texas and Florida are the only states whose suits name Altman personally.

The sequence shows how state consumer-protection law is becoming a de facto AI regulatory tool. Rather than waiting for a single federal AI statute, state attorneys general are using long-standing deception and child-online-safety statutes to seek injunctions that reshape product design and release practices.

Background: why consumer-protection law fits this fight

Deceptive trade practices statutes were written for misleading advertising and unfair business practices, not for frontier AI models. Their appeal to enforcers is breadth: if a company misrepresented a product's safety, a court can order changes to the product, its marketing, and its distribution — not merely a fine. The SCOPE Act adds a minors-focused layer, covering data practices and parental controls for young users.

For AI companies, the theory raises questions courts have only begun to answer. What counts as a deceptive safety claim about a general-purpose chatbot, whose outputs vary by prompt? How should a court weigh warnings and safety features against marketing language? And can a state court supervise the release of a nationally distributed model without creating conflicting obligations across states? Texas's case will test the first steps of those questions.

Why it matters for founders and investors

For founders, the suit warns that safety claims are legal commitments, not marketing copy. Every statement that a product is "safe" or "suitable for children" can become an exhibit. Founders should be able to show, in documents, what testing was done, what risks were found, what mitigations shipped, and what warnings users actually saw. Age-gating and parental controls should be built early, not retrofitted under a court order, and launch plans should tolerate a safety-review delay.

For investors, diligence should now cover litigation exposure by state, the exact wording of safety marketing, minors' access controls, and whether the company has independent safety evaluations it can point to in court. The personal naming of a chief executive also raises the stakes for leadership accountability, insurance, and governance. A company with strong technology but weak safety documentation carries regulatory risk that can delay launches and raise legal costs.

Angeled.com connects founders with investors.

MLex, October 24, 2026: Texas lawsuit seeks court-approved safety reviews before OpenAI releases new models.
AIStockWire, October 23, 2026: Texas sues OpenAI and Sam Altman over ChatGPT safety claims; OpenAI lawsuit tracker.
Zot News / Particle News, October 23, 2026: reporting on the Texas filing in Collin County district court.

Read Next